The long term objective of this project is to explore how the metabolism of glucose and other secretagogues stimulates insulin secretion. The studies outlined in this proposal will expand on our observation that mitochondrial glycerol phosphate dehydrogenase (mGPD), which participates via the glycerol phosphate shuttle in the reoxidation of NADH produced by glycolysis, is more abundant in the pancreatic beta cell than in other tissues. This suggests that the glycerol phosphate shuttle is important for insulin secretion. In support of this idea are recent reports that mGPD activity is low in pancreatic islets in rodent models of non-insulin dependent diabetes mellitus and in several humans with this disease. We recently cloned the rat mGPD cDNA and found that the deduced sequence contains a region with homology to the calcium-binding sites of calmodulin. This may explain the long-known calcium activation of mGPD and suggests that calcium interacts directly with mGPD to activate it during glucose stimulation. We have isolated rat islet and human mGPD cDNAs and human genomic clones, and have produced a potent mGPD antibody in order to examine the role of mGPD in insulin release and diabetes. As there are no known potent specific inhibitors of mGPD, we will attempt to obtain direct evidence that the glycerol phosphate shuttle is important for insulin secretion by experiments designed to reduce mGPD levels with ribozyme and antisense strategies in beta cell lines. Transgenic mice with decreased beta cell mGPD will be produced by pronuclear injection of an insulin promoter-mGPD-ribozyme construct. An mGPD null mouse will be produced by targeted gene disruption. Insulin secretion induced by glucose will be compared with that induced by compounds not requiring shuttle activity. mGPD enzyme activity, protein, and mRNA levels, and glycerol phosphate shuttle activity will be measured. Experiments are described to clone and characterize the mGPD gene, including mapping intron-exon splice junctions and looking for tissue (islet) specific cDNAs. The mGPD gene is increased by thyroid hormone in tissues where its activity is low, such as liver, whereas in tissues, such as the pancreatic islet, where mGPD is extremely abundant, thyroid hormone has no effect. Promoter studies will investigate this apparent high level of constitutive expression in the islet, as well as the thyroid hormone responsiveness in liver. Important elements in the promoter and 5' flanking domain will be identified and transacting factors that interact with these regions will be studied. Research will also be undertaken to study the functional sites in the mGPD protein by in vitro mutagenesis of the FAD site, the calcium binding site and the putative glycerol phosphate binding domain. Collaborative studies of mGPD in families with non-insulin dependent diabetes and in animal models of this disease will be continued. Information gained from these studies will increase the understanding of glucose-stimulated insulin release and non- insulin dependent diabetes.